SUSE

October 2011

 

 

 

 

Connect

Sales form

Have a question?

Need support?

Want to talk?

Contact us

Give us your thoughts

Read Past Issues

Introducing the SUSE Geeko Gazette

Welcome to the Geeko Gazette, a quarterly newsletter for SUSE Linux Enterprise users. The Geeko Gazette will provide you with the latest information on technical solutions, product updates and training, as well as hardware and software certifications. Our mission is to ensure that you have the resources and know-how to get the most out of your SUSE Linux Enterprise environment, improve the efficiency of your data center and make your job easier.

SUSE Tips and Tricks

2-3+4+7+0=10

Improve Interoperability Between SUSE Linux Enterprise 10 and Windows 7 Active Directory

By Matthias Eckermann, Senior Product Manager

This equation is mathematically correct. However, it also demonstrates how one can easily improve interoperability between SUSE Linux Enterprise 10 and Windows 7 Active Directory.

3+4=7

With Windows 7, Microsoft not only changed its naming scheme once again (remember Windows 2000, 2008, XP and Vista?) but also added enhanced authentication mechanisms to Active Directory.

3.(0|2) = (2|3)

Up to and including Samba 3.0, Samba has been licensed under General Public License v2 (GPLv2). But, with the release of Samba 3.2, the software is now licensed under GPLv3.

10

Package licenses in SUSE Linux Enterprise do not change during the lifecycle of a major release. Our customers and partners demand that we keep this rule. Looking at our current releases:

  • SUSE Linux Enterprise 10 was released before GPLv3 was available. Thus, it ships with Samba 3.0 and hundreds of tools and packages licensed under the GPLv2 and other licenses.
  • SUSE Linux Enterprise 11 ships with packages under both GPLv2 and GPLv3, among other licenses, and includes Samba 3.4.

3

Unfortunately, older versions of Samba 3 (3.0 and before) do not work with the authentication mechanism in Windows 7 and will fail. Workarounds exist, but Windows administrators do not like them and are hesitant to apply this type of backward compatibility. And I don't blame them!

4

Have you heard about Samba 4? It has great authentication features, which can easily work together with Windows 7 and its Active Directory implementation. The problem is that Samba 4 has not been released.

3+4=3.4

Luckily, however, the Samba project team devised a way to leverage some of the capabilities of Samba 4 in a Samba 3 environment. The Samba team's solution supports a combined build, where the code is not intermixed. This enables Samba 3 and Samba 4 to share some of the same infrastructure components and gain better Windows 7 integration. No, this is not called Samba 7, but Samba 3.4.

3+4 != 10

Reading the above, it may not be apparent that there is a direct way to provide full interoperability between SUSE Linux Enterprise 10 and Windows 7 Active Directory. But certainly, you expect us to find a way.

2

And we did. Our solution is to provide two Samba versions. We include Samba 3.0 with SUSE Linux Enterprise 10. We are also delivering Samba 3.4 separately, and it is ready for use with SUSE Linux Enterprise 10. Starting in December 2010, we began offering separate update channels in the Novell Customer Center: “SLES10-GPLv3-Extras” for the server and “SLED10-GPLv3-Extras” for the desktop, respectively. These channels contain the same Samba versions that were shipped in SUSE Linux Enterprise 11 Service Pack 1, recompiled and appropriately packaged for SUSE Linux Enterprise 10. The client and server YaST2 Samba modules have been adopted and updated to work with either Samba version (i.e., 3.0 or 3.4).

2+3=5

Here are four steps we recommend to install the latest Samba version on your SUSE Linux Enterprise 10 system:

1. Ensure the system is registered in the Novell Customer Center

2. rug sub SLES10-GPLv3-Extras # register new channel

3. rug ref # refresh

4. rug in samba-gplv3 # install the GPLv3 Samba

5. Enjoy!

For more information see TID 7007836


The Short Guide to Fast and Affordable Virtualization

Virtualize Your Data Center with SUSE Linux Enterprise Server and SUSE Manager

By Joachim Werner, Senior Product Manager

Want to set up, run and maintain an affordable virtualized data center? All you need are SUSE Linux Enterprise Server with KVM or Xen virtualization and a SUSE Manager server with the Management and Provisioning modules enabled for all physical servers you want to use as virtualization hosts. Virtualization management has never been more integrated with the management and monitoring of the rest of the operating system stack.

Interested? Want to get started? Just follow these simple steps:

1. Install the SUSE Manager server. This is straightforward, as SUSE Manager comes as a pre-integrated appliance that you can deploy within minutes on a physical server or as a virtual instance.

2. Install a minimal server system with either the KVM or Xen hypervisor (you have the choice, both are fully supported by SUSE) onto all the servers you want to use for virtualization and register them to SUSE Manager. With the built-in PXE server you can do this directly from SUSE Manager without touching the physical server.

3. Use AutoYaST, which is integrated into SUSE Manager, and the SUSE Manager configuration management features to help you specify the setup of those machines in detail. To add additional host servers to the virtualized environment, you can just reuse the same template.

4. After the virtualization hosts are registered to SUSE Manager you can create virtual server workloads directly from the SUSE Manager web interface and deploy them to one of the virtualization hosts. With the optional SUSE Manager Monitoring module you can even monitor the health and availability of both your physical hosts and your virtual guest systems from that one web interface. AutoYaST and SUSE Manager's advanced templating mechanisms can create reusable configurations that allow you to add new virtual servers in a matter of minutes.

Finished! With those four steps, you have created a virtual environment using only SUSE Linux Enterprise Server and SUSE Manager. But what about ongoing management of my virtualized environment, you ask? Read on.

With SUSE Manager, you can group physical and virtual servers by location, workload, or any other categories you choose, while keeping them up to date with the SUSE Manager patch management capabilities. This is where the combination of SUSE Linux Enterprise Server and SUSE Manager really excels. No other system or virtualization management console allows you to accomplish all of those tasks, from provisioning new hosts and virtual machines to keeping all the software packages up to date, using a single system and interface.

If your environment then grows beyond what one administrator can handle, or if you are running shared services, SUSE Manager allows you to create multiple logical organizations and assign servers and software subscriptions to those organizations according to their needs. You can then add users to those organizations that are allowed to perform only the tasks the users are authorized for.

If your data center grows beyond what mere mortals can sanely handle with even the most comfortable web interface, you can start automating repetitive tasks beyond what the templating and configuration management features of SUSE Manager allow you to do with the comprehensive SUSE Manager API or command line client. For example, you can create a new organization, assign administrators to the organization, and add the servers and subscriptions they are supposed to be using in a single scripted transaction. The API also allows you to integrate your virtualized environment into your existing tools for automation as well as reporting and compliance.

SUSE Manager and SUSE Linux Enterprise Server with KVM or Xen are all you need to set up a highly automated and manageable virtualized data center, at an unmatched cost. Give it a try! Both SUSE Manager and SUSE Linux Enterprise Server are available for free evaluation.


The File System of the Future

An Introduction to btrfs
By Matthias Eckermann, Senior Product Manager

You may have heard or read that btrfs (pronounce it "better fs") is the file system of the future in Linux because of features such as:

  • Increased data integrity based on checksums on data and metadata
  • Improved scalability and extensibility
  • Powerful file system snapshot capabilities
  • Support for "Copy on Write"
  • Integrated volume management

Lucky for you, the future is now. SUSE included btrfs as a technical preview in SUSE Linux Enterprise Server 11 SP1 and we will fully support it in the upcoming release of SUSE Linux Enterprise Server 11 SP2. As part of SUSE Linux Enterprise Server 11 SP2, SUSE will provide the tools to maximize the capabilities of btrfs. For example, SUSE engineers have worked with the openSUSE® community to develop command line tools and YaST2 integration for managing btrfs snapshots. This demonstrates, once again, the power of open source development to speed the delivery of meaningful innovation for our customers.

You should test btrfs now in SUSE Linux Enterprise Server 11 SP1 so you'll be ready when it is fully supported in SUSE Linux Enterprise Server 11 SP2.


Is It Time for a Cluster Stack Health Check?

Ensure Your Clusters are Optimized for Maximum Performance

By Kai Dupke, Senior Product Manager

Linux clustering solutions are becoming a higher priority for many enterprises, especially as their use of Linux for mission-critical services continues to grow. Customers that rely on continuous stability and performance from their Linux clusters cannot afford suboptimal cluster configurations.

To help our customers ensure their cluster configurations are running at peak performance, SUSE has teamed with the high availability experts at LINBIT to offer a cluster stack health check service.

The LINBIT Cluster Stack Health Check is an extensive check-up for Linux high availability clustering solutions, such as the SUSE Linux Enterprise High Availability Extension. LINBIT Professional Services provides a thorough examination of an organization's cluster stack and a detailed best practice report with recommended enhancements to help customers get the most from their high availability cluster configurations.

Are your clusters running at peak performance, particularly the mission-critical components? SUSE customers with a SUSE Linux Enterprise Server subscription can get LINBIT's highly effective Cluster Stack Health Check for a special price until December 31, 2011. Click here for more information.


Building and Deploying Mainframe Images Made Simple and Fast

SUSE Studio Advanced Edition takes the Pain out of Workload Deployments for IBM System z

By Wolfgang Engel, Senior Engineer

In early 2000, IBM, based on client demand, expanded use of open source software and enabled Linux to run on IBM mainframe computers. With the mainframe's reliability, availability and serviceability coupled with a suite of robust virtualization capabilities, customers saw the mainframe as an ideal platform on which to run multiple Linux servers to make more efficient use of their computing resources. Since 2000, when SUSE Linux Enterprise Server became the first commercially-available enterprise Linux distribution for IBM System z mainframes, SUSE has delivered Linux-based mainframe solutions that customers across all industries can deploy to expand their critical applications and workloads.

The recently released SUSE Studio Advanced Edition builds on that experience by providing a new and simple way to create, maintain and deploy IBM System z workloads by automating routine, labor-intensive tasks, which increases efficiencies and further reduces complexity and costs.

SUSE Studio is designed to help enterprise organizations build, configure and maintain cloud-enabled application images, reducing the complexity, maintenance and support costs of software deployments. SUSE Studio Advanced Edition provides a bridge between x86 microprocessor architectures and mainframes by simplifying the creation, testing, maintenance and deployment of software applications on the mainframe. Now, customers running hundreds of mission-critical Linux images on mainframes can save time by leveraging an easy-to-use interface for building mainframe workloads.

Building and Deploying System z Images

But, where do you begin? To build System z images you need a set-up that consists of an x86 Linux server running SUSE Studio and an S/390 Linux machine that builds and creates the System z image.

Once the set-up is finished you select "Create new appliance" on SUSE Studio's web interface and choose the base template "System z". This will create a new base template for System z. Using that template you can now select packages and customize your images with Studio. Once you have configured the image you can build it by selecting the "build tab". That was fast. But, how do you deploy your image?

There are two disk formats for System z that you are able to build:

DASD Deployment Steps

If you built an image in DASD format, you can deploy it on a real DASD on your System z. You can do this by either using an existing Linux machine and copying the image to a new DASD with dd or by using z/VM and it's copying program CMSDDR.

For example, if you are using an existing Linux machine for this task you have to activate the destination DASD by using the command "dasd_configure <your_device_address> 1". This attaches the destination DASD to the Linux system and enables the access due to a device entry e.g. /dev/dasdf.

Then you have to copy the image you created with studio to the existing Linux machine, either by using scp, ftp or even wget directly from your onsite location.

When the copy process has finished you need to dd the image to the DASD like this "dd if=<your_image_name> of=/dev/dasdf bs=4k". After that make sure all data has been written to the disk by calling "sync".

The last step is to detach the destination DASD by using the command "dasd_configure <your_device_address> 0".

After that you are able to ipl the destination DASD from either a z/VM machine or LPAR.

zFCP Deployment Steps

If you choose to build an image in zFCP format, the steps for writing the image on the disk with dd is the same as with DASD with only the steps for activating the scsi-disk changing.

There are two simple methods for activating your target scsi-disk. You can use YaST2 → "Hardware" → "Zfcp" or you can use two scripts. The first script enables you to bring your zFCP adapter online:

"zfcp_host_configure <your_zfcp_adapter_device_address> 1"

The second script allows you to access the disk and make it visible to your system:

"zfcp_disk_configure <your_zfcp_adapter_device_address> <wwpn_of_your_storage> <disk-lun> 1"

If you use a scsi-disk (e.g. /dev/sda) you will need to use dd to copy the image to the disk:

"dd if=<your_image_name> of=/dev/sda bs=4k ; sync"

Please make sure that you deactivate your scsi-disk after the image is written to the disk:

"zfcp_disk_configure <your_zfcp_adapter_device_address> <wwpn_of_your_storage> <disk_lun> 0"

After these steps you can ipl your scsi-disk either using z/VM or LPAR.

SUSE Spotlight

Influencing the Evolution of Linux to Address Customer Needs

An Interview with Engineering Fellow, Greg Kroah-Hartman

SUSE Linux Enterprise gives customers a secure, supported distribution for running their applications and workloads. To ensure the SUSE Linux Enterprise distribution continues to address the needs of our customers, SUSE employs a team of engineers that look to evolve the distribution in a way that responds to customer demand, interest and feedback. One such engineer is Greg Kroah-Hartman, a SUSE Fellow in the SUSE Labs division.

While Greg has been with SUSE for five years on the kernel team, he has been a contributor to and maintainer of the Linux distribution for more than 10 years. Greg’s initial participation in the community was simply as a hobby, but he is now one of the top 10 contributors to the Linux kernel. He authored the book Linux Kernel in a Nutshell, which explains how to build your own Linux kernel, and he co-authored the book Linux Device Drivers, which instructs programmers and device manufacturers how to develop Linux device drivers. One of Greg’s main responsibilities is device enablement, working with hardware manufacturers to ensure their devices work properly with the SUSE Linux Enterprise product line.

“We have strong relationships with a large number of companies that want to work with Linux, such as HP, IBM, Intel and others,” says Greg. “These relationships are a great benefit to our customers because it takes out the middleman. Customers know that if they encounter a problem, they can go straight to the source of who owns, creates and maintains the code,” Greg adds. “Those relationships also allow us to assist the manufacturers with their hardware enablement and help ensure they continue to contribute to and remain involved in how Linux evolves.”

In fact, Greg indicates that one of the most interesting aspects of his job is to introduce companies to the Linux kernel community and help them become participating members of the community that will take ownership for the driver code for their individual products. As part of this responsibility, Greg has documented the details of how somebody can become involved in the Linux kernel development process here: http://lxr.linux.no/#linux+v2.6.38/Documentation/HOWTO.

Greg explains that another reason people become involved in the kernel community is that it gives them the opportunity to influence the evolution of Linux, an evolution that has been moving at a rapid pace. The time between major kernel releases can vary, but releases usually occur between eight and 12 weeks. When asked how the kernel community keeps up with the massive rate of change and growth, Greg answers, “It’s our great skill.” While he says that with a smile, he adds, “Seriously, we have a very good hierarchy of developers and maintainers, very good tools, and we constantly evaluate our processes to ensure they’re working well.”

A significant number of those skilled developers in the community work with Greg as part of the SUSE team. “That’s a great benefit that SUSE customers enjoy,” he says. “SUSE engineers know this code. That allows us to integrate new features our customers need, as well as make sure it works properly on all the new hardware that customers have and will have in the future.”

By having engineers like Greg and other kernel community contributors on the SUSE team, customers get direct input on how the Linux kernel evolves; including helping define the priorities on fixes and new features. “As we become aware of customer needs and desires, we can directly influence kernel evolution so those needs are addressed,” Greg says. “That’s the main goal of our participation, as well as getting other companies to participate in the process, to make sure our customers’ current and future needs can be met. And because we’re part of that community with an excellent staff of engineers, we have the ability to do exactly that.”

Cloud Corner

SUSE Expands Relationships with Amazon and IBM

Easier Deployment and More Choice in the Public Cloud

By Peter Chadwick

In the last edition of “The Geeko Gazette,” we talked about the availability of Amazon EC2 running SUSE Linux Enterprise Server. Judging from the level of usage we have seen since launching the offering, a number of you are already using it to develop, test and deploy virtual appliances in EC2. Since then, we have launched two enhancements to make Amazon EC2 running SUSE Linux Enterprise Server an even more compelling choice for your cloud workloads: one-click deployment from SUSE Studio™ and Cluster Instances.

One-Click Deployment

When we initially launched the availability of SUSE Linux Enterprise Server in the Amazon EC2 cloud, we expanded SUSE Studio with the ability to create Amazon EC2 images as an option during the image build process. Initially, these images had to be downloaded from SUSE Studio and manually uploaded to EC2. With the recent release of SUSE Studio 1.2, you can now launch these images directly into Amazon EC2.

All you need to take advantage of this exciting feature is an Amazon EC2 account. After configuring your image in SUSE Studio, you select "Amazon EC2 image" as the target deployment platform. After the build is complete, you will see an option to "Upload to EC2". The first time through, you're prompted to provide AWS credentials, which you can store for subsequent uploads. You are then prompted to select the EC2 instance type and region where you want the Amazon Machine Image (AMI) loaded.

After uploading the image, you can choose whether to immediately launch a new instance of it. As part of the upload process, SUSE Studio automatically creates an Amazon EC2 key pair to provide secure access to a running instance. The file that contains the key pair can be downloaded by selecting "Show connection information" after the upload completes. Once the instance is running, this key pair can be used to access the instance via SSH.

After it is deployed, that image can even be managed along with your other on-premise and cloud instances using SUSE Manager. The combination of SUSE Studio and SUSE Manager enable you to balance the flexibility of the public cloud with the control you seek for your data center.

Cluster Computing Instances

The Amazon EC2 Cluster Compute and Cluster GPU instance types are specifically designed to combine high compute performance with high performance network capabilities to meet the needs of High Performance Computing (HPC) applications. Unique to Cluster Compute and Cluster GPU instances is the ability to group them into clusters of instances for use with HPC applications. This is particularly valuable for those applications that rely on protocols such as Message Passing Interface (MPI) for tightly coupled inter-node communication. These instance types benefit from the same on-demand flexibility as all other Amazon EC2 instance types and provide an excellent vehicle for developing and testing HPC applications.

SUSE Linux Enterprise Server has long been recognized as one of the best performing operating systems for HPC and is the operating system running six of the 10 fastest supercomputers in use today. Now customers can bring the same capabilities to Amazon EC2 to take advantage of both the cc1 and cg1 instances. Because of the differences with standard EC2 images, cluster instance images cannot be built using SUSE Studio, but instead need to be configured using standard SUSE Linux Enterprise tools when the instance is running. The cg1 instances, which include the Nvidia Tesla graphics processor, will automatically install the necessary drivers.

IBM SmartCloud Enterprise

SUSE is also seeking to establish more public cloud partnerships around the globe to ensure that enterprises have the flexibility to deploy a solid and supported enterprise Linux offering with their preferred cloud provider. One example of our continued expansion is that we have added support for SUSE Linux Enterprise Server in the recently announced IBM SmartCloud Enterprise.

IBM Smart Cloud Enterprise is an agile cloud infrastructure as a service (IaaS) designed to provide rapid access to security-rich, enterprise-class virtual server environments. It is well suited for development and test activities and other dynamic workloads. Ideal for both IT and application development teams, the IBM Cloud delivers cloud-based services, systems and software to meet the needs of your business.

SUSE Linux Enterprise is available for purchase as virtual machine instances, and also forms the base operating system for a wide range of predefined software images that combine IBM middleware such as IBM WebSphere, IBM DB2 Express and IBM Lotus Web Content Management. Virtual machine images are available in 32-bit or 64-bit instance sizes and are optionally available with IBM Premium Support. To get started just click here.

Training and Support

Who Says a Good Thing Doesn't Last Forever?

SUSE Provides UNIX-like Lifecycle through Long Term Service Pack Support Program
by Douglas Jarvis, Product Marketing Manager

Ten years is not quite forever, but it is still a long time. With our Long Term Service Pack Support (LTSS) program, 10 years is how long you can now run each major version of SUSE Linux Enterprise Server, while continuing to receive security updates and bug fixes. We recently expanded the LTSS program to include three years of extended support for SUSE Linux Enterprise Server on top of our standard seven years of general support. The addition of extended support provides customers with a UNIX-like 10-year lifecycle for each generation of SUSE Linux Enterprise Server at a much lower cost of ownership.

Our updated LTSS program also includes five years of support for SUSE Linux Enterprise Server service packs. Typically, general support for our service packs is available for two years, but with LTSS you can now obtain support for up to three additional years. This change enables you to upgrade to the next service pack at your own pace, or skip a service pack altogether and move to a later one.

And LTSS isn't just for industry-standard hardware. The Linux workloads you have running on your mainframe are exceptionally important and that is why we now offer LTSS globally for SUSE Linux Enterprise Server for System z. So whether you are running SUSE Linux Enterprise Server on System z or industry-standard hardware, LTSS reduces your risk by ensuring that the configurations for your mission-critical workloads can be maintained over an extended period of time.

It may not be forever, but the expansion of SUSE's LTSS program ensures that your workloads can continue to run in a stable, secure and supported environment for a long time. See how LTSS can help you take advantage of the latest technology at the pace that makes sense for your business.

Certifications

The Hardware and Software Support You Need

Updates on Hardware and Software Certifications

By Peter Chadwick, Senior Product Manager

The SUSE dedicated hardware and software certification teams have been hard at work ensuring that you get the support for the applications you need and for the systems you want to run them on. In the first half of 2011 alone, our hardware team completed 1,152 certifications. This included certifications of the following:

Acer–AC100, AW370 Compute Module F1

Cisco–UCS B230 M2, UCS B440 M2

Dell–PowerEdge R810, PowerEdge R910, PowerEdge 6800, PowerEdge C6145

HP–BL620c G7, BL680c G7, DL980 G7

IBM–x3850 X5, x3690 X5, x3755 M3, HS22, HS22V, HX5

Oracle–Sun Fire X4470 M2, Sun Fire X4800 M2

Search for more hardware certified on SUSE Linux Enterprise.

Additionally, in the first half of 2011, our independent software vendor (ISV) team added 1,492 certified applications for SUSE Linux Enterprise 10 and 11. As of June 30, 2011, SUSE maintained 6,103 application certifications for SUSE Linux Enterprise 10 and 11 from nearly 1,600 ISVs. These totals contain 606 certified Oracle applications, including the latest versions of Oracle Database, Oracle Fusion, E-Business Suite, PeopleSoft, Siebel, JD Edwards, Financial Applications, and Project Portfolio Management.

If you know of an application that is not supported, let us know. Our ISV team will work with the application provider to get the applications you need certified on SUSE Linux Enterprise. We're committed to making everything run on SUSE Linux Enterprise so you don't have to worry about it. See a complete listing of all certified applications.

In this Issue

SUSE Tips and Tricks

2-3+4+7+0=10

The Short Guide to Fast and Affordable Virtualization

The File System of the Future

Is It Time for a Cluster Stack Health Check?

Building and Deploying Mainframe Images Made Simple and Fast

SUSE Spotlight

Influencing the Evolution of Linux to Address Customer Needs

Cloud Corner

SUSE Expands Relationships with Amazon and IBM

Training and Support

Who Says a Good Thing Doesn't Last Forever?

Certifications

The Hardware and Software Support You Need